1 / 9

Intelligent Routing Algorithm Using Antnet

The delay in the transporting packets or data from one point to the other has become a big problem in communication network. This can be surmounted by characterizing a data network with a view in finding out the throughput performance, modeling a dynamic routing algorithm that provides paths that change dynamically in response to network traffic and congestion, thereby increasing network performance because data travel less congested paths, simulating the intelligence routing algorithm using Ant net that has properties like learning, reasoning and decision making with respect to packet transmission in a data network using MATLAB SIMULINK as a tool and comparing the performance of the model to existing routing algorithm Aneke Israel Chinagolum | Chineke Amaechi Hyacenth | Udeh Chukwuma Callistus. W "Intelligent Routing Algorithm Using Antnet" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-3 | Issue-1 , December 2018, URL: https://www.ijtsrd.com/papers/ijtsrd18990.pdf Paper URL: http://www.ijtsrd.com/computer-science/artificial-intelligence/18990/intelligent-routing-algorithm-using-antnet/aneke-israel-chinagolum<br>

Download Presentation

Intelligent Routing Algorithm Using Antnet

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. International Journal of Trend in Scientific Research and Development (IJTSRD) International Open Access Journal | www.ijtsrd.com ISSN No: 2456 - 6470 | Volume - 3 | Issue – 1 | Nov – Dec 2018 Intelligent Routing Algorithm Using Antnet Aneke Israel Chinagolum1, Chineke Amaechi Hyacenth2, Udeh Chukwuma Callistus. W3 1Computer Science Department, Institute of Management and Technology Enugu, Enugu, Nigeria 2Computer Engineering Department, Micheal Okpara University of Agriculture Umudike, Abia, Nigeria 3ICT Department, Enugu State University of Science and Technology Teaching Hospital, Enugu, Nigeria ABSTRACT The delay in the transporting packets or data from one point to the other has become a big problem in communication network. This can be surmounted by characterizing a data network with a view in finding out the throughput performance, modeling a dynamic routing algorithm that provides paths that change dynamically in response to network traffic and congestion, thereby increasing network performance because data travel less congested paths, simulating the intelligence routing algorithm using Ant net; that has properties like learning, reasoning and decision making with respect to packet transmission in a data network using MATLAB/SIMULINK as a tool and comparing the performance of the model to existing routing algorithm Keywords: Routing, intelligent, ANTNET 1.INTRODUCTION: Large-scale, wide area data networks are a part of today’s global communication Networks such as the Internet have become an integral medium of information transfer, ranging from personal communication to electronic commerce and entertainment. The importance of such networks will only increase as the electronic world becomes more prevalent. The basic function of a data network is very simple: delivering data from one network node to another. Achieving this goal requires many network components, including physical computers and links, signaling protocols between computers, and data packaging protocols. This thesis addresses one such component, routing; the process that logically connects network nodes by calculating paths between nodes so that data sent by one node traverses the calculated path to its destination. Although many algorithms in graph and operational research literature calculate paths between nodes, the challenge in developing network routing algorithms is in dealing with the scale and distribution of the physical network. Because typical wide area networks have nodes on the order of tens of thousands, routing algorithms must be scalable. In addition, routing algorithms must be able to calculate paths in a distributed manner due to the global and distributive nature of physical networks. Moreover, because of the actual physical network, routing algorithms need to cope with events such as physical component failures and recalculate paths whenever such events occur. Finally, routing algorithms need to calculate paths to allow nodes to achieve high throughput performance. In general, routing algorithms view a network as a weighted graph, where network links are represented as graph edges and network routers as graph vertices. Network routers are network nodes that execute routing algorithms and ensure that data travel the calculated paths. In the weighted graph, the assignment of edge weights depends on the specific routing algorithm; typically, the assignment reflects the latency and bandwidth of the link (Atul, 1999). After a routing algorithm makes these link cost assignments, it then computes paths between nodes. Thus, the specific routing algorithm that routers execute determines the paths that data will travel in the network. Routing algorithms in today’s Internet base their implementations on the static metric single shortest pathrouting model. Single shortest path means that routing algorithms provide, at any given time, the least-cost path between nodes. Static metric refers to link cost assignments which are based on static infrastructure. @ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 3 | Issue – 1 | Nov-Dec 2018 Page: 306

  2. International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470 3.To simulate the intelligence routing algorithm using Antnet; that has properties like learning, reasoning and decision making with respect to packet transmission in a data network using MATLAB/SIMULINK as a tool 4.To compare the performance of the model to existing routing algorithm 3.Methodology To characterize a data network with a view in finding out the throughput performance Table 1 : DATA COLLECTION ON PACKET LOSS DUE TO CONGESTION WITH DATE OF DATA COLLECTION: 1th TO 8th of FEBUARY, 2018 Packet Transferred Data 12.00 AM 1.00 AM 2.00 AM 3.00 AM 4.00 AM 5.00 AM 6.00 AM 7.00 AM 8.00 AM 9.00 AM 10.00 AM 11.00 AM To calculate the efficiency in 12.00 AM Recall Efficiency = 100% x (transferred-retransmitted)/ Transferred Efficiency = 100 x (30 – 25)/ 25 Efficiency = 100 x 5/25 Efficiency = 100/5 = 20 Then, to calculate the network loss in 12.00AM Network loss or packet loss = 100 – efficiency Network loss = 100 -20 = 80 = 0.8 properties of a link, such as its bandwidth or latency. As shown later, the main drawback of this model is that static metric shortest paths do not always provide good network performance. 1.Problem Statement The general problem of routing packets in transport networks is in itself nothing new. The shortest path problem is a basic part of graph theory, with multiple solutions and countless applications in different fields. Previous implementations suffer two problems: routing instability and high routing overheads. Routing instability occurs when paths are constantly being recomputed and do not stabilize. This instability degrades network performance and increases the probability of network congestion and failures. The second problem is high routing overheads. The overheads are in terms of the CPU cycles and messages required for path re-computation. So, one of the fast changing areas in internetwork routing is the application of intelligence into it. Among the major challenges that exist in network, the coexistence of multiple infrastructure systems and different models are some of the prime issues. The existing infrastructure lacks the intelligent awareness and coordination of numerous components and applications running on the network and it has opened up various complexities. Also, the coordination with different models needs a good understanding of the services provided by the layers. These two criteria had led the researchers to think of making the network nodes more intelligent. It is reasonable to say that intelligence in the nodes must include properties like learning, reasoning and decision making. To establish accurate results for optimal path, a network must be made aware of its environment. This can be realized by helping the network to learn, think and remember and hence making it intelligent. 2.Research Objective This research is focused on proposing an intelligent routing algorithm for data network using path vector technique. Its main objective includes: 1.To characterize a data network with a view in finding out the throughput performance. 2.To model a dynamic routing algorithm that provides paths that change dynamically in response to network traffic and congestion, thereby increasing network performance because data travel less congested paths. Packet Packet Loss Time Retransmitted 30 25 0.8 28 24 0.833 26 20 0.7 26 18 0.556 24 16 0.5 24 14 0.2858 22 12 0.167 22 11 0.167 0.182 22 11 20 12 0.2 18 13 0.6154 18 14 0.7143 @ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 3 | Issue – 1 | Nov-Dec 2018 Page: 307

  3. International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470 To calculate the efficiency in 1.00 AM Efficiency = 100% x (transmitted– retransmitted)/ Transmitted Efficiency = 100 x (28 – 24)/24 Efficiency = 100 x 4 /24 Efficiency =16.7 = 0.167l To calculate the network loss Network loss = 100 – 16.7 Network loss = 83.3 Network loss = 0.833 To calculate the efficiency in 2.00 AM Efficiency = 100% x (26 – 20)/20 Efficiency = 100 x 6/20 = 30 Network loss = 100 – 30 = 70 = 0.7 To calculate the efficiency in 3.00AM Efficiency = 100% X (26 -18)/18 Efficiency = 100 x 8 /18 Efficiency = 800/18 = 44.4 Network loss = 100 – 44.4 Network loss = 55.6 = 0.556 To find the Efficiency at 4.00 AM Efficiency = 100% X (24 - 16)/16 Efficiency = 800/16= 50 Network loss = 100 – 50 = 50 = 0.5 To find the Efficiency at 5.00 AM Efficiency = 100 x (24 -14)/14 Efficiency = 1000/14 = 71.42 Network loss = 100 – 71.42 Network loss = 28.58 = 0.2858 To calculate the efficiency at 6.00 AM Efficiency = 100 x (22 – 12)/12 Efficiency = 1000/12 Efficiency = 83.3 Network loss = 100 – 83.3 Network loss = 16.7 = 0.167 To calculate the efficiency at 7.00 AM Efficiency = 100 x (22 -12)/12 Efficiency = 1000/12 Efficiency =83.3 Network loss = 100 – 83.3 Network loss = 16.7 = 0.167 To calculate the efficiency at 8.00 AM Efficiency =100 x (20 – 11)/11 Efficiency = 900 /11 Efficiency Network loss = 100 -81.8 = 18.2 Network loss = 0.182 To calculate the efficiency at 9.00 AM Efficiency = 100 x (22 – 10)/10 Efficiency = 800/10 = 80 Network loss = 100 – 80 Network loss = 20 = 0.2 To calculate the efficiency at 9.00 AM Efficiency = 100 x (20 -12)/12 Efficiency = 800 /12 = 66.67 Network loss = 100 – 66.67 Network loss = 33.3 = 0.333 To calculate the efficiency at 10.00 AM Efficiency = 100 x (18 – 13)/13 Efficiency = 500/13 Efficiency = 38.46 Network loss = 100 – 38.46 Network loss = 61.54 = 0.6154 To calculate the efficiency at 11.00 AM Efficiency = 100 x (18 -14)/14 Efficiency = 400/14 = 28.57 Network loss = 100 -28.57 Network loss = 71.43 = 0.7143 Table 2: DATA COLLECTION ON PACKET LOSS DUE TO CONGESTION PER HOUR Packet loss experienced hourly 0.8 0.833 0.7 0.556 0.5 0.2858 0.167 0.167 0.182 0.2 0.6154 0.7143 To find the congestion at 12.00 AM The mathematical model for congestion control in improving optimized real time monitoring and evaluation of GSM quality of service using intelligent agent is as shown in equation 3.2 = 81.8 Congestion experienced hourly 2.04 1.96 2.33 2.94 3.266 5.67 9.78 9.78 8.97 8.17 4.6 2.29 @ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 3 | Issue – 1 | Nov-Dec 2018 Page: 308

  4. International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470 To find the congestion at 12.00 AM L = 8/ 3W^2 ----------------------------------------------1 Where L is packet loss W is the network congestion Then, make W the subject formula in equation 1 The mathematical model for congestion in the network is as shown in equation 2 W = Square root of 8/3W^2-----------------------------2 To find the network congestion in 12.00 AM W1 = square root 8/ 3 x 0.82 W1 = square root 8 /3 x 0.64 = 8/1.92 W1 =√ 4.17 W1 =2.04 Congestion in 1.00 AM W2 = square root 8/3 x 0.8332 = 8 / 3 x 0.8332 = 8/2.081667 W2 =√ 3.84 W2 = 1.96 Congestion in 2.00 AM W3 = Square root of 8/ 3 x 0.72 = 8/1.47 W3 = square root of 8/ 1.47 = 5.442 W3 = √5.442 W3 = 2.33 Congestion in 3.00 AM W4 = Square root of 8/ 3 x0.5562 W4 = square root of 8/ 0.927 W4 = √8.63 W4 = 2.94 W4 = 2.94 Congestion in 4 A.M W5 = square root of 8/3 x 0.52 W5 = Square root of 8/ 0.75 =10.67 W5 = √10.67 W5 = 3.266 Congestion in 5 A.M W6 = Square root 8 / 3 x 0.28582 W6 = Square root 8 / 0.25 W6 = √32 W6 = 5.67 Congestion in 6 A.M W7 = Square root 8/ 3 x 0.1672 W7 = Square root 8/ 0.0837 W7 = √95.57 W7 = 9.78 Congestion in 7A.M W8 = 8/ 3 x 0.1672 W8 =8/0.0837 W8 = 8/0.0837 W8 = √95.57 W8 = 9.78 Congestion in 8A.M W8 = 8/ 3 x 0.1822 W8 = 8/0.099372 W8 = 80.5 W8 = √80.5 W8 =8.97 Congestion in 9 A.M W9 = 8/3 x 0.22 W9 = 8/0.12 W9 = √66.7 W9 = 8.167 Congestion in 10 A.M W10 = 8/3 x 0.61542 W10 = 8 /0.378717 W10 =√ 21.12 W10 = 4.6 Congestion in 11 A.M W11 = 8/3 x 0.71432 W11 = 8/1.53 W11 = 5.23 W11 = √5.23 W 11 = 2.29 To determine an ideal bit error rate convenient for the characterized network Taking into consideration the worst case scenario, the linear relationship between BER and packet error rate (PER) is expressed as: PER= 8 × BER ×MTU ×66/64--------------------------3 Where the MTU is the maximum transmission unit, and using the Ethernet standards it is set to 1500 bytes for the simulations and then the MTU is increased to improve performance. A conversion from 8 bits to 1 byte is shown, Recall 1 byte = 8bits 1500bytes = 8 x 1500 = 12000bits MTU = 12000bits PER is packet loss and BER is bit error rate To evaluate the bit error rate in 12 A.M when the packet loss is 0.8 @ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 3 | Issue – 1 | Nov-Dec 2018 Page: 309

  5. International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470 Bit error rate in 9 A.M BER10 = 0.2/9900 = 0.000020bits Bit error rate in 10A.M BER11 = 0.6154/9900 = 0.000062bits Bit error rate in 11A.M BER12 = 0.7143/9900 = 0.000072bits Table 3: DATA COLLECTION ON PACKET LOSS DUE TO CONGESTION PER HOUR WITH BIT ERROR RATE Packet Loss Experienced Hourly Hourly 0.8 2.04 0.833 1.96 0.7 2.33 0.556 2.94 0.5 3.266 0.2858 5.67 0.167 9.78 0.167 9.78 0.182 8.97 0.2 8.17 0.6154 4.6 0.7143 2.29 To model a dynamic routing algorithm that provides paths that change dynamically in response to network traffic and congestion, thereby increasing network performance because data travel less congested paths Make BER the subject formula in equation 3.3 BER1 = PER/8 × MTU × 1.03125 ---------------------4 BER1 = 0.8/8 x 12000 x1.03125 BER1 = 0.8/9900 BER1 = 0.000081bits To find the bit error rate in 1 A.M BER2 = 0.833/9900 BER2 = 0.000084bits Bit error rate in 2 A.M BER3 = o.7/9900 BER3 = 0.000071bits Bit error rate in 3 A.M BER4 = 0.556/9900 BER4 = 0.0000562bits Bit error rate in 4A.M BER5 = 0.5/9900 BER5 = 0.000051 bits Bit error rate in 5A.M BER6 = 0.2858/9900 BER6 = 0.000030 bits Bit error rate in 6A.M BER7 = 0.167/9900 BER7 = 0.000017bits Bit error rate in 7A.M BER8 = 0.167/9900 BER8 = 0.000017bits Bit error rate in 8A.M BER9 = 0.182/9900 = 0.000018bits Congestion Experienced Bit Error Rate 0.000081bits 0.000084bits 0.000071bits 0.0000562bits 0.000051bits 0.000030bits 0.000017bits 0.000017bits 0.000018bits 0.000020bits 0.000062bits 0.000072bits Fig 1: Designed model for a dynamic routing algorithm that provides paths that change dynamically in response to network traffic and congestion @ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 3 | Issue – 1 | Nov-Dec 2018 Page: 310

  6. International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470 Fig 1 shows designed model for a dynamic routing algorithm that provides paths that change dynamically in response to network traffic and congestion. In fig 1 the design was done in mat lab software with the necessary blocks like channel, scope, and transmitter. The empirical data collected from the GLO network were inserted inside the designed model. To simulate the intelligence routing algorithm using Antnet; that has properties like learning, reasoning and decision making with respect to packet transmission in a data network using MATLAB/SIMULINK as a tool Fig 2 Simulated result of development of improving of throughput in data network without using intelligent routing. Fig 2 shows simulated result of development of improving of throughput in data network without using intelligent routing. In fig 2 the simulated result is a high bit error rate coupled with congestion that make the network not to function effectively well. Fig 3 Simulated result of development of improving of throughput in data network using intelligent routing. Fig 3 shows simulated result of development of improving of throughput in data network using intelligent routing. Fig 3 when simulated reduced both bit error rate and congestion concurrently thereby making the network to be free. To compare the performance of the model to existing routing algorithm Fig 3 gives a better result by reducing congestion and bit error rate than figure 2. @ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 6 | Sep-Oct 2018 Page: 311

  7. International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470 4.Result Analysis Table 4: CONVENTIONAL CONGESTION VS NUMBER OF USERS CONVENTIONAL CONGESTION 2.06 1.96 2.33 2.94 3.27 5.67 9.78 9.78 8.97 8.17 4.60 2.29 NUMBER OF USERS 1 2 3 4 5 6 7 8 9 10 11 12 Fig 4CONVENTIONAL CONGESTION VS NUMBER OF USERS Fig 4 shows conventional congestion vs number of users. The lowest coordination of congestion vs number of users is (1.96, 2) while the highest coordination of congestion vs number of users (8.97,8). Table 5: CONVENTIONAL CONGESTION VS NUMBER OF USERS USING INTELLIGENT ROUTING CONVENTIONAL CONGESTION USING INTELLIGENT ROUTING 1.442 1.386 1.648 2.079 2.312 4.009 6.916 6.916 6.343 5.777 5.253 1.619 NUMBER OF USERS 1 2 3 4 5 6 7 8 9 10 11 12 @ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 3 | Issue – 1 | Nov-Dec 2018 Page: 312

  8. International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470 Fig 5 CONVENTIONAL CONGESTION VS NUMBER OF USERS USING INTELLIGENT ROUTING Fig 5 shows conventional congestion vs number of users using intelligent routing. The highest coordination of congestion vs number of users is (6.916,7) and (6.916,7). The lowest coordination is (1.442, 2). This shows that there will be free communication network that its congestion is free. Table 6: COMPARING CONVENTIONAL CONGESTION AND CONVETIONAL CONGESTION VS NUMBER OF USERS WITH INTELLIGENT ROUTING ALGORITHM CONVENTIONAL CONGESTION USING INTELLIGENT ROUTING 2.06 1.442 1.96 1.386 2.33 1.648 2.94 2.079 3.27 2.312 5.67 4.009 9.78 6.916 9.78 6.916 8.97 6.343 8.17 5.777 4.60 5.253 2.29 1.619 CONVENTIONAL CONGESTION NUMBER OF USERS 1 2 3 4 5 6 7 8 9 10 11 12 Fig 6 COMPARING CONVENTIONAL CONGESTION AND CONVETIONAL @ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 3 | Issue – 1 | Nov-Dec 2018 Page: 313

  9. International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470 CONGESTION VS NUMBER OF USERS WITH INTELLIGENT ROUTING ALGORITHM. Fig 6 shows comparing conventional congestion and conventional congestion vs number of users with intelligent routing algorithm. The highest coordination of conventional approach and intelligent routing approach are(9.78,7) and (.6.916.7). This shows that there is free network in communication network when intelligent routing is used. 5.Conclusion The manner by which free communication network is easily noticed in our communication network has arisen as a result of high bit error rate, that constitutes interference and congestion which has led to economic degradation among the subscribers that in bark in business and the country at large can be overcome by development of intelligent routing algorithm for the improvement of throughput in data network. This can be overcome in this manner. Characterizing a data network with a view in finding out the throughput performance. Modeling a dynamic routing algorithm that provides paths that change dynamically in response to network traffic and congestion, thereby increasing network performance because data travel less congested paths. simulating the intelligence routing algorithm using Antnet; that has properties like learning, reasoning and decision making with respect to packet transmission in a MATLAB/SIMULINK as a tool and comparing the performance of the model to existing routing algorithm Reference 1.Elizabeth E. Peterson routing algorithm using heretical technique 2017 2.Joseph, Jack routing algorithm using flat based technique 2014 3.Leonard Anderson routing algorithm using proportional integral (PI) 2013 4.Theodore Samuel routing algorithm using proportional derivative (PD)2016 data network using @ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 6 | Sep-Oct 2018 Page: 314

More Related